CODES AND STANDARDS ENHANCEMENT REPORT

Duct Sealing Requirements upon HVACor Duct-System Replacement: LightCommercial Buildings

Methodology 5

Results 6Reproduction or distribution of the whole or any part of the contents ofthis document without the express written permission of PG&E isprohibited. Neither PG&E nor any of its employees makes any warranty,express or implied, or assumes any legal liability of responsibility for theaccuracy, completeness, or usefulness of any data, information, method,policy, product or process disclosed in this document, or represents thatits use will not infringe any privately-owned rights, including but notlimited to patents, trademarks or copyrights ..

Bibliography and Other Research 9

Appendices 11

This Code Change Proposal encompasses two changes to the existing California Building Energy EfficiencyStandards: 1) a requirement that duct systems be sealed and tested at the time that an air-conditioner, heat-pump, orfurnace is installed in certain existing light commercial buildings, and 2) a requirement that new or replacement ductsystems in that same class of buildings meet the same insulation levels as ducts in similarly designed new buildings(i.e. R-8), as well as be sealed and tested. The proposed requirements are triggered either by installation of a furnace,a packaged indoor/outdoor unit, or replacement or installation of an air handler, cooling or heating coil, or furnaceheat exchanger, or by installation of a new or replacement duct system in an existing structure. In order for a ductsystem to be submitted to these requirements, it must meet three conditions: 1) it must be part of a system thatserves 5000 ft2 or less of conditioned floor area, 2) it must be connected to a constant-volume, single-zone HVACequipment, and 3) the ducts must be located either outside the envelope of the building (e.g. on top of the roof), orbelow a roof with inadequate insulation.

If triggered, the requirement is that the ducts be sealed such that their measured leakage is less than 15% of fan flowfor the supply and return ductwork combined, including the HVAC equipment cabinet and plenums, or that the ductsystem pass a visual smoke test and have their leakage be reduced by 60% relative to the condition before sealing.

All installations that include duct sealing shall be self-certified by the installing contractor, including leakage testing,and shall be submitted to a verification program by a third-party special inspector or building inspector. Theverification program shall follow a set of procedures that are functionally equivalent to those currently used forresidential new-construction duct sealing requirements.

The existence of the proposed requirements for duct sealing shall be publicized on the CEC and other public interestweb sites, as well as through campaigns such as the Flex-Your-Power campaign. These requirements should also bepublicized on a voluntary basis by HV AC equipment distributors.

California, through the Title-24 California Building Energy Efficiency Standards, has made tight ducts an integralpart of low-rise residential new construction in California, as well as a compliance alternative for new lightcommercial buildings. These changes to the Standards, enacted over the past 5 years, were based upon favorablecost-benefit analyses. The expected benefits of sealing ducts in light commercial buildings are even larger than inresidential buildings, especially in existing buildings. This is due to higher, more-consistent daytime baselineconsumption in commercial buildings, as well as due to higher initial leakage levels in existing light commercialbuildings. The options of insulating and/or coating roof surfaces, rather than sealing ducts, were included because ofthe fact that sealing may not always be practical, and the proposed alternatives are shown to be equivalent to ductsealing in terms of energy savings and/or peak demand reduction produced.

Another reason to address energy efficiency in existing buildings is that more air conditioners and furnaces areinstalled in existing buildings as compared to new construction, both in California and nationwide. In California, alittle more than 60% of this equipment is installed in existing buildings, or in other words, 50% more equipmentgoes into existing buildings as opposed to new construction each year. The proposed change makes the standardsapply to all duct systems, no matter what type ofHV AC installation is involved-newly constructed buildings, ,additions, and alterations (replacements).

m PG&E Code Proposals (DRAFT - Do not Quote)

Duct sealing also can improve indoor air quality and safety, principally by reducing entry of outdoor pollutants intothe living-space, including reduced ozone entry during smog alerts, and reduced entry of air from dusty ceiling-plenum spaces.

On a statewide level, there are approximately 140,000 furnaces, air conditioners, or heat pumps installed in existinglight commercial buildings each year (Source ARI). According to research performed by Lawrence BerkeleyNational Laboratory, 60-65% of the duct systems in the existing buildings where this equipment is installed wouldmeet the eligibility criteria for application of the proposed changes. In addition, according to data collected on 350light commercial duct systems for Southern California Edison, more than 85% of the duct systems tested in thetarget population of light commercial buildings merited sealing, with the average combined leakage on the supplyand return sides of the system adding up to 36% offan flow. Assuming that all of the eligible HV AC equipmentinstallations that merit duct sealing (0.625*0.85*140,000=74,000) actually receive duct sealing or an alternative atthe time of equipment installation, the estimated annual statewide savings added each year that the proposed changeis in effect are: 46 GWh, 0.5 million therms, and 35 MW. The assumptions behind these estimates are included inthe Appendix.

This change does not have any adverse environmental impacts. The only materials used are commonly usedmaterials: building sealants, tapes, and fiberglass insulation. One potential environmental impact would be toincrease the rate at which asbestos ductwork in the existing building stock is removed. The degree to which this willbe the case is uncertain, as many contractors may choose the alternative measures when they encounter asbestosductwork. It is also possible that this change may reduce combustion safety problems in light commercial buildings,at least in the case of buildings with vented ceiling-plenum spaces. Research performed by LBNL (Delp et al. 1998)suggests that supply ducts leak 25% offan flow, whereas field work performed for Southern California Edison(Modera and Proctor, 2002) indicates total leakage of 36% of fan flow, suggesting that most systems are supplyleakage dominated. Sealing supply-leakage-dominated systems should reduce building depressurization, therebyreducing combustion-safety problems.

The proposed change is a Prescriptive Measure that must be met whenever a light commercial furnace, evaporatorcoil, or packaged unit is installed in an existing light commercial building, or whenever a duct system is added to orreplaced in an existing light commercial building.

The proposed change encompasses small changes to both the Standards and the ACM. In both cases, the changes arerelatively minor, as outlined in the section below.

The principal suppliers of this measure are the HVAC contractors who normally install HVAC equipment inexisting buildings, roofing contractors, or building insulation contractors who service existing light commercialcustomers. The methodology and supplies required by these contractors for accomplishing the proposed changes areprovided by a range of suppliers. These include duct sealant manufacturers, manufacturers of duct leakage testequipment, companies that supply training to HV AC technicians, and manufacturers of insulation and cool-roofproducts. There are multiple suppliers in each of these categories, and there already exist hundreds of contractorsand technicians who have been trained to produce verified tight ducts through utility training programs. Many ofthese contractors already own the equipment required to verify duct tightness. Sealing supplies are available frommultiple manufacturers.

There exists adequate capacity to meet the expected increase in demand for training, duct sealants, sealingequipment and duct insulation. In addition to the training staff and facilities at utilities, there are several companiesthat sell duct-improvement training and diagnostic/sealing equipment for contractors and technicians, including

m PG&E Code Proposals (DRAFT - Do not Quote)

Advanced Energy, Carrier-Aeroseal, Comfort Institute/Retrotec, The Energy Conservatory, andHoneywell/Enalysis. The means by which training and sealing equipment are distributed include direct sales ofequipment, complete diagnostic/sales systems, and franchises that provide one-stop shopping for training,sealing/diagnostic equipment and diagnostic/sales tools.

The baseline condition that this measure is attempting to change is simple replacement ofHV AC equipment withoutany change to the energy efficiency of the duct system to which it is connected. Currently, less than 10% of ductsystems receive verified sealing or additional insulation at the time of equipment replacement, and the currentstandard does not require these measures. The other significant aspect of the baseline condition is that an unknownnumber ofHV AC replacements are performed without a building permit being issued.

The costs for this change are time and materials for sealing and leakage testing of existing ducts in existingdwellings, or the incremental costs of sealing and testing replacement duct systems combined with the incrementalcost of using R-8 instead ofR-4.2 replacement ducts. Sealing costs are estimated to be $930, based upon observed$150/ton costs from the SCE duct sealing project (Modera and Proctor 2002), combined with a $30/systemincremental cost associated with third-party verification services (DEER 200 I). It should be noted that these costsare for stand-alone duct sealing without any legislative requirement. Costs should be lower when duct sealing is astandard part of every equipment installation, both because it becomes part of a bigger job (i.e. combined travel andtransaction costs), and because there is no selling involved. Duct replacement with verified-tight R-8 ducts isestimated to cost $1052, based upon a combination of sealing and testing costs plus $122 in incremental ductinsulation costs (Source: Owens-Corning Fiberglas). The duct sealing costs include self-verification measurementcosts on every job, and third-party verification costs that correspond to current cost offield verification on one inevery five jobs. Maintenance costs are not an issue for these technologies.

For all cost-effectiveness analyses, the useful life of duct sealing and insulation was assumed to be 30 years,consistent with the values used for new construction within the current Title-24 Standards and ACM.

The proposed change requires a third-party verification program. The testing protocol for the duct-sealing alternativeis similar to that used for new construction, which is already outlined in Appendix G ofthe non-residential ACMManual. In addition to the testing protocol, the third-party verification mechanisms also need to be specified. In thiscase, the proposed change will utilize the HERS rater mechanism, including the use of Third Party Quality ControlPrograms as outlined in section 7 of the non-residential ACM.

The key issue with respect to enforcement of this change in the Standards is the potentially significant fraction ofHV AC equipment that is installed without building permits. This proposal does not address that issue directly, butrather proposes several alternatives for helping to increase the use of permits, and therefore the degree ofenforcement of the proposed change. These alternatives include publicity ofthe change on CEC and other publicinterest web sites, publicity through campaigns such as the Flex-Your-Power campaign, and publicity on a voluntarybasis by HV AC equipment distributors.

The cost effectiveness of the proposed change was evaluated based upon savings calculated using the DOE-2.2program, which was checked against the calculation procedures currently in Appendix G of the Non-residentialACM. The energy consumption levels were generated by applying the DOE2.2 program to a prototype building withlower-performance windows and higher lighting power as compared to the new-construction prototype used toanalyze similar measures in new construction. Duct sealing cost values were obtained from observed contractorcosts from the 2001-2002 Southern California Edison project to reduce demand through light commercial ductsealing (Modera and Proctor, 2002), combined with costs for on-site third-party verification applied to one in everyfive installations (DEER, 200 I). The marginal cost of going from R-4 to R-8 at duct replacement was calculatedm PG&E Code Proposals (DRAFT- Do not Quote)

from data supplied by a large insulation manufacturer. Cost effectiveness was evaluated for five climate zones forduct sealing and for R-8 tight duct replacements, while energy consumption values were computed for the variousalternatives to duct sealing. The average results for the five climate zones (3, 6,10,12 and 14) are presented below.These results are based upon assuming a 60%/40% split between vented-plenum ceiling-only-insulation andunvented-plenum ceiling&roof-insulation. All runs assumed the use of an economizer and 6% minimum outdoor air,however the results do not change significantly without economizers. All of analyses were performed using an R-19ceiling, which, based upon the way that the Standards language is worded, actually provides a conservative estimateof the savings for insulated roofs. The reason is that when the roof insulation must always be less than or equal tothe ceiling insulation, the lower the insulation level on the ceiling, the higher the overall energy use, resulting inhigher absolute energy savings. Climate-by-climate results for sealing and alternatives are summarized in theappendix.

Measure Consumer Discounted Lifetime Time Dependent ValuationCost Savings (5-climate average)

Duct Sealing and Testing $ 930 $2,591

Duct Replacement w/tight R-8 $ 1052 $ 2,958

This change does not require the use of any analysis tools, as what is proposed is a set of prescriptive requirements.In situations where the performance methodology is used in lieu of prescriptive alternatives, the calculationprocedures are those currently in Appendix G of the non-residential ACM, potentially modified as per another codechange proposal for ducts in new construction.

This change will not have any significant impacts on other measures. This code change proposal was developed inconjunction with a similar duct-sealing proposal for new construction, and thus was designed to be consistent with,and to complement that proposal. This code change proposal was also designed to be as consistent as possible withthe duct sealing requirements for existing residential systems.

Methodology

The lack of requirements for duct efficiency improvements at the time of HVAC replacement has been recognizedas an area for improvement since 1998, and was identified by the California Energy Commission as a topic for thisproceeding during the AB970 effort. Research consisted of reviewing the standards language to identify a strategy toincorporate replacement duct sealing or the equivalent into the standard. The methodology used to analyze this codechange proposal is based upon savings estimates calculated using the DOE-2.2 program, which was checked againstthe calculation procedures currently in Appendix G ofthe Non-residential ACM.

To estimate the cost effectiveness of duct tightening, a series of simulation studies were undertaken. First, a simple"box" prototype model was developed to test the capabilities and evaluate the response of the DOE-2.2 program toseveral duct efficiency and operating condition assumptions. The eQUEST program was used to develop the basicDOE-2.2 input file. Manual changes were made to the input file to complete the analysis. A description of thesimple box model is shown below:

m PG&E Code Proposals (DRAFT - Do not Quote)

Model Parameter ValueShape Rectangular, 50x40Conditioned floor area 2000 SFNo Floors I

Floor to ceiling 9ftPlenum ht 3ftWindow/wall ratio 20%WindowSC Varies by climate zoneWindow U-Value Varies by climate zoneExterior wall const 8 in. concrete tilt-up construction insulatedExt wall R-Value Varies by climate zoneInfiltration rate 0.3 ACH in occupied zone, varies in atticRoof construction Built-up roof over plywood deckRoof absorptivity and emissivity Abs = 0.8; emiss = 0.9Ceiling construction Acoustic tileLighting power density 2 W/SFEquipment power density 0.5 W/SFOperating schedule 7 am - 6 pmM-FNo. People II

Outdoor air 15 CFM/person (minimum 6% of flow)HVAC system PSZSize 6 ton (CZ 3,6) 10 ton (CZ 10,12,14)CFM 2100 CFM (CZ 3,6) 3500 CFM (CZ 10,12,14)SHR @ ARl conditions 0.7EER 8.5Fan power 0.28 W/CFMSupply duct surface area 27% of floor area, per ACMDuct leakage 36% total leakage; evenly split between supply

and return (18% supply, 18% return) for leakycase, 15% total leakage for tight case

Duct insulation R-value R-4.2, with an air film resistance of 0.7 addedto account for external and internal air filmresistance. R-8 evaluated for some runs.

Return leak from outside air 0%Return system type ducted

m PG&E Code Proposals (DRAFT- Do not Quote)

The proposed changes to the Standards below are predicated on duct sealing and insulation requirements beingapplied to new construction also, as proposed in a companion code change proposal. Should such requirements notbe enacted for new construction, the changes proposed herein would need to be modified so as to make themapplicable only to alterations.

The proposed changes differ from those originally presented at the July 18, 2002 workshop. The modifications to theproposal were made: 1) to respond to comments received relative to the original proposal, and 2) to maintainconsistency with the residential code and ACM changes. Specific changes to the proposal include: 1) increasing themaximum allowable leakage level from 10% to 15% offan flow, 2) allowing 60% sealing of the existing leakageplus smoke testing as an alternative, and 3) allowing standard roof insulation as an alternative to duct sealing. Theprincipal impact ofthese changes is to make it easier to meet the sealing requirements by means of manual ductsealing. In addition, the trigger for the duct-sealing requirement was modified so as to reduce the probability ofrepairing rather than replacing the HV AC equipment to avoid the duct-sealing requirement. This change was madeto address concerns raised during and subsequent to the July 18, 2002 workshop.

(k) Air Distribution System Duct Leakage Sealing. A.I duct systems shall be sealed to a leakage rate not toexceed 6% of the total fan flow if the duct system:

The leakage rate shall be confirmed through field verification and diagnostic testing, in accordance with proceduresset forth in the Nonresidential ACM Manual.

EXCEPTION TO Section 144(k) 3 B: Where the roof by Itself meets the requirements of 143(a) 1 C.

BQ. New space-conditioning systems or components other than space conditioning ducts shall meet therequirements of Section 144; and

D. New space-conditiomng ducts shall meet the reguirements of section 124; and If they meet the criterIa of section144(k) 1.,2., and 3., be sealed as confirmed through field verification and diagnostic testing in accordance withprocedures for sealing of existing duct systems as specified in the Nonresidential ACM manual, to meet one of thefollowing reguirements:

~&--------~&~ PG&E Code Proposals (DRAFT - Do not Quote)

ii The duct leakage shall be reduced by more than 60% relative to the leakage prior to the equipment havingbeen replaced, and a smoke test shall demonstrate that all accessible leaks have been sealed.

E. When a space-conditioning system is installed or replaced (including replacement or installation of the airhandler, cooling or heating coil, or furnace heat exchanger), the existing duct system, if it meets the criteria ofsection I44(k) 1.,2., and 3., shall be sealed, as confirmed through field verification and diagnostic testing inaccordance with procedures for sealing of existing duct systems as specified in the Non-Residential ACM manual,to one ofthe requirements of section I49(b) 1 D; and

EXCEPTION 1 to Section 149 (b) 1 E: Installation or replacement of only the outdoor condensing unit of a splitsystem air conditioner or heat pump.

EXCEPTION 2 to Section 149 (b) 1 E: Buildings altered so that the duct system no longer meets the criteria ofSection I44(k) I., 2., and 3_

EXCEPTION 3 to Section 149 (b) 1 E: Duct systems that are documented to have been previously sealed asconfirmed through field verification and diagnostic testing in accordance with procedures in the Non-ResidentialACM manual.

The modifications shall made to Section 7 are incorporated into the ACM per the new-construction ductperformance proposal.The following text shall be added to Section 4 of Appendix G:

The apparatus for verifying best-effort duct sealing shall also include means for generating non-toxic visual smokefor identifying leaks in accessible portions of the duct system. This apparatus may consist of a theatrical smokegenerator, or a smoke bomb. In both cases, adequate smoke must be generated to assure that any accessible leakswill have to emit visibly identifiable smoke.

When the diagnostic leakage test is performed for existmg duct systems pursuant to Section 149 (b)l.D. or Section149 (b)1.E. of Title 24, Part 6, the procedures described in Section 4.3.8.2.3 shall be followed.

m PG&E Code Proposals (DRAFT - Do not Quote)

For existing duct systems, the requirements of Section 149 (b) 1.c. or Section 149 (b)1.D. of Title 24, Part 6 shall beconsidered satisfied if one of the following two conditions is met:

1. the measured total duct leakage is less than 15% of the total fan flow, where the duct leakage shall bedetermined pursuant to the procedures in section 4.3.8.2.1, and the total fan flow shall be determinedpursuant to section 4.3.7,

2. the duct system passes the followmg two tests on ever): job: a) duct leakage has been reduced by more than60% relative to the leakage prior to the equipment ha\>ingbeen replaced, determined pursuant to section4.3.8.2.1 before and after sealing the duct system, AND b) a smoke test is performed according to section4.3.8.2.3.1 in the presence ofa third-party tester to show that best efforts have been made to seal accessibleleaks.

The objective of the smoke test is to confirm that best efforts have been made to seal all accessible leaks, and shallbe comprised of the following steps:

I) Injection ot either theatrical or smoke-bomb smoke into a fan pressurization device that ismaintaining duct pressure between 25 and 50 Pa relative to the duct surroundings, with all grillesand registers (as well as any intentional outdoor air intakes) in the duct system sealed.

11 r he only smoke emanating from the system can be clear!) identified as coming from theHVAC equipment rather thar the ducts

The Non-Residential Manual will also need to be modified to make it consistent with the proposed changes to theStandard and the Non-Residential ACM. The residential manual should also include additional alternatives to ductsealing.

Bibliography and Other Research

ASHRAE Standard 152P: Method of Test for Determining the Energy Efficiency of Residential ThermalDistribution Systems under Seasonal and Design Conditions. Second Public Review Draft 08/01.

W. W. Delp, N. Matson, D. J. Dickerhoff, D. Wang, R. C. Diamond, M. P. Modera "Field Investigation of DuctSystem Performance in California Light Commercial Buildings", ASHRAE Trans. 104(11) 1998, June 1998.

m PG&E Code Proposals (DRAFT - Do not Quote)

W. W. Delp, N. Matson, D. J. Dickerhoff, D. Wang, R. C. Diamond, M. P. Modera "Field Investigation of DuctSystem Performance in California Light Commercial Buildings (Round II)." Proceedings of ACEEE Summer Study,Pacific Grove, CA, August 1998.

M.P. Modera, J. Proctor, "Combining Duct Sealing and Refrigerant Charge Testing to Reduce Peak ElectricityDemand in Southern California", Final Project Report for Southern California Edison, July 2002.

PG&E sponsored this proposal under direction of Pat Eilert. The contractor for this project is the Heshong-MahoneGroup.

The report was prepared by Mark Modera of Modera Consulting Engineers and Peter Jacobs of Architectural EnergyCorporation. The authors would like to acknowledge the sponsorship by PG&E ofthis report and code changeproposal, and in particular the guidance provided by Marshall Hunt as the PG&E project manager. The authorswould also like to thank the following persons who provided key input to the project:

John McHugh of the Herschong-Mahone Group. Mr. McHugh collaborated on the DOE 2.2 modeling procedures, aswell as on the development and analysis of code language alternatives.

Robert Scott (CHEERS) Mr. Scott provided a perspective on the ability and desire of CHEERS raters to support thethird-party testing requirements of this proposal

m PG&E Code Proposals (DRAFT - Do not Quote)

Appendix

• 140,000 furnaces and/or air conditioners installed in existing light commercial buildings each year;

• 62.5% of the duct systems located above ceiling insulation;

• 85% of installations leak enough to trigger sealing requirement;

• all triggered installations are addressed (i.e., 74,000 efficiency improvements);

• all installations have central Ale;

• all installations have base-case efficiency levels of AFUE-80 and EER-8.5;

• elevated savings associated with heat pumps or electric furnaces are excluded;

• savings from duct replacements or outdoor ductwork not included in statewide total;

• 37.5% of existing buildings have insulated ceilings and vented attics, 25% have insulated ceilings and roofswith no ventilation, and 37.5% have insulated roofs and no venting;

• improvements split uniformly between the five climate zones simulated;

• 30-year life for duct sealing and insulation measures;

• $0.129 value per KBtu for TDV savings estimates (Time Dependent Valuation (TDV) EconomicsMethodology, John McHugh, HMG, http://www.h-m-g.com/TDV/index.htm)

• Peak demand reduction calculated based upon maximum draw determined by DOE 2.2 for different options.

m PG&E Code Proposals (DRAFT - Do not Quote)

Figure A.1 Duct Sealing Savings for Various Climates and Roof Configurations(R-19 ceiling)

$5,000OCZ3

$4,500 OCZ6$4,000 .CZ10

$3,500 IlliICZ12t/) EllCZ14g> $3,000> • AveragenJ $2,500en> $2,000Ct-

$1,500

$1,000

$500

m PG&E Code Proposals (DRAFT - Do not Quote)

Vented R-ORoof Unvented R-O Unvented R-10 Unvented R-19Roof Roof Roof

$600$500

t/) $400C)c:oS;

$300cuUJ> $200cI-

$100$0

R-B vs R4.2 tight ducts R-B vs R4.2 tight ducts R-B vs R4.2 tight ductsfor unvented R19/RO for unvented R19/R10 for unvented R19/R19

OCZ3

IIIiICZ6

.CZ10

lIDCZ12

mCZ14

.Average

• Duct sealing costs of $150/ton, augmented to include third-party field verification cost of $30/system, whichcorresponds to an effective field verification rate of one in five systems installed;

• Marginal costs for duct replacement include the cost of sealing, as well as increased material cost associatedwith installing R-8 instead ofR-4.2 ducts (Material Cost Data obtained from Dave Ware, Owens ComingFiberglas).

m PG&E Code Proposals (DRAFT - Do not Quote)